Hacker News new | past | comments | ask | show | jobs | submit login
America’s housing system still has not been properly reformed (economist.com)
164 points by martincmartin on Aug 18, 2016 | hide | past | favorite | 144 comments



I'd agree that the US housing and derivatives markets are in a much better position than in 2007-2008. Atleast people who are getting mortgages now have some form of documentation, and risk manages are being asked to do their jobs.

The crisis now is that, and this is just my opinion, low rates are here to stay. with the amount of money in the system, people are being priced out of homes and as time goes on the amount by which they are priced out in the big cities is only going to grow.

As an aside...

I'm not a lawyer but I've heard people say My cousin Vinny is the most accurate court room movie ever. In a similar vein, I'd nominate the Big Short as one of the more accurate wall street movies ever made.

It showed the lone wolf who not only saw the coming crash, this was far and away the easiest part, but figured out how to profit form it by getting the banks to write him a CDS.

The two smart kids working out of their garage that were smart enough to succeed despite all their mistakes.

The hedge fund who did an incredible amount of due diligence, going to properties, knocking on doors, talking to people inside the industry to put together the pieces that no one else could.

The wall street sales guy who sold the CDS knowing that it might bankrupt his company but figured he'd make a killing on his commissions long before that happened.

And the most accurate part of the movie was that it showed the anguish that each of these people went through. They'd seen something no one else had, been right about it, had the event happen and even then everyone else continued to plug their ears and pretend nothing was going on while they continued to bleed money each month. It really did a good job of showing that you can be right about everything and still not make money if the rest of the market continues to act insane.

In 2007 there really was a sense that people knew things were way out of hand but everyone was thinking, "If I can just get one or two more years on the bonus cycle, I'll be able to get out". To be fair, this type of thinking probably ins't going away anytime soon.


The MIT Blackjack team wrote a good piece on the housing crash, paying particular attention to the market distortion caused by short-term rewards.

https://yourmortgageoryourlife.wordpress.com/2009/01/15/the-...


Honest question: how many of them reverted to mean afterward [0], that is, thought they saw other things coming, bet on them and were wrong about as often as everyone else?

John Paulson wasn't featured in the movie, but fit the bill. But I don't know about those in the book/film.

[0] https://en.wikipedia.org/wiki/Regression_toward_the_mean


If I remember correctly one of them closed down his fund because he was stressed and tired of his clients always second guessing him through the whole thing. Then begin to second guess him again on his next decisions despite being right the first time and bringing in hundreds of millions of dollars in profit.


The Christian Bale character closed his fund to withdrawals from investors as the shit was hitting the fan but the big banks were trying to keep it under control, thus preventing his default swaps from paying off. He was losing money as the banks tried to stall the inevitable.

A couple months later he made everyone who was suing him to get their money millions of dollars.


I believe he is investing in water now if the end of the film is correct. Let's hope he isn't right twice.


Well, there are interesting reports out there about water and the companies involved with that commodity.


Could you elaborate?


My bad, missed your response.

There are the occasional stories about companies that draw water from a local community for their own purposes, usually for retail, when that community would likely be better off without that relationship. In some cases there are shenigans involved and complaints over the contracts.


Margin call with Kevin Spacey and Zachary Quinto is another excellent portrayal of the crash. It's set inside a fictional unnamed bank that's securitizing these mortgages when thinks go pear shaped.


Nice comment. Food for thought.

How big was the risk that they took ? Surely they knew that the market was insane. But would they have lost a lot of money if the market was insane for a long time ? I saw the movie but I don't remember a lot of the details. I was just trying to follow.


IIRC from reading the book a year or two ago, one of them's bosses were hounding him for the premiums he was paying on the insurance-like contracts he bought, which was something like $25m/year. When the crash happened they bought in ~$1b. Had there not been a crash he would have got nothing (the contracts would never have paid out) and that $25m/year would have been money down the drain.


Shorting is a risk. Shorting with big money is a big risk.

In this case though, my impression is that certain people who actually looked at the data knew it would happen. Not like "I think it will happen", it was "it will happen sometime in this month of this year" kind of thing. Because they knew when those stupid ARMs would kick in with their adjustable rates on top of the obvious number of failed mortgages. The risk wasn't that it would not happen, it was that high level people in finance and government would take steps to keep things moving along which would prevent them from cashing in. There was a great deal of fraud throughout that whole thing propping up the failed system.


"take steps" kind of sugar coats it, from the film it sounds like the banks strongarmed the ratings companies to lie about the quality of the derivatives as the ARMs folded.


That was my impression from sources outside the film as well. Like it was stated in the film, "if we rate them lower they'll go down the street to our competitor".


The other risk was that the entity on the other side of the bet would be wiped out in the crash and not be able to pay out.


That was part of the problem in delaying them from cashing out.


It was not difficult to see the housing bubble would burst. The problem is you can't make money knowing that without getting the timing mostly right. That's the hard part.


> Shorting is a risk. Shorting with big money is a big risk.

Calling it "shorting" suggests the downside was unlimited (as it is when shorting equities). That wasn't the case for this trade.


Yes, but unlike a normal short their loss was limited. They had (IIRC) a 7 year contract to pay $25million/year unless things went bust in which case they get paid $1 billion. So their worse case was a $175 million loss, their best case was a billion gain. If you have the money the potential return on invest is great.

I'm about half way through the book now, but I don't have it handy at the moment to look up the real numbers though.


By that measure the returns on a lottery ticket is magnitudes better.

The big question is the likelihood of the gamble paying off. For a lotto ticket that's easy to figure. Trying to time the collapse of a bubble? If that was easy bubbles wouldn't exist. Maybe it's implied that 7 years was clearly more than enough time, but that's only clear in hindsight.

Now some of the winners of the collapse seemed to have had a firm grasp of the situation. But how would _you_, as an investor, know that? How do you measure that?

Warren Buffett can write all the books in the world for as long as he likes, but successfully communicating his knowledge, his strategies, his intuition, has proven fairly difficult. It's probably not even possible because Warren Buffett probably doesn't 100% understand his strategy. Likewise for other consistently successful people. If they were capable of communicating it, there'd be many more people like them.

It's like the old adage, if you code to the best of your ability, by definition you're not capable of debugging it.


> The big question is the likelihood of the gamble paying off. For a lotto ticket that's easy to figure.

I agree. The gamble is tricky.

> Now some of the winners of the collapse seemed to have had a firm grasp of the situation. But how would _you_, as an investor, know that? How do you measure that?

Read all the data. Defaults of 4% are expected in the ideal market. Their bet only needed defaults to reach 5% to pay out. Those who started in the game carefully selected from the sub prime mortgages the ones most likely to default after 2 years: the interest only no documentation loans. The real question is why were they able to short the ones more likely to default for the same cost as the ones less likely to default. Those risks were all upfront.

Note that the methods these people used is obvious even with foresight: read the data on the investments, it was all there. The only hard part was the data was 100 pages of fine print. This data was was there and honest, if I gave any investor this data and forced them to read it they would arrive at the same results. Again though, reading the data is a hard job and so most of those who should read it didn't, they just assumed everything was as it was before not noticing the the numbers of no documentation loans was going from few to a majority.

If there was any luck involved it was only that the investors who made money had the idea to one day not read a stock 10-k report and instead read a bond report. There were signs that these guys saw that others didn't, but once they had the idea to look at the data it really is a no brainier.

The only thing I didn't have is enough money to invest (minimum investment is $30 million and that was raised to $100 million quickly), the patience to wait 2 years for it to pay off, the luck to think of the bond market (if I was looking for a great stock in China at the time I might have found something good, but not this great investment in bonds), and the willingness to wade though all these finance documents looking for any that are worth investing on. The last is by far the most important: I find the data boring and so I don't want to analyze it.

Note that the subprime bubble is over. You cannot pull that off today in subprime. Or at least I don't think you can: I haven't looked at the data. Every 10-20 years some opportunity comes up. 3/4ths the battle is the idea to look in the right place, the other 1/4th is to actually understand all the data. There might be something now, there might not. I don't have a crystal ball.

Even if I did, I don't have enough money to ride out the market. As more than one person has said, "the market can remain irrational longer than you can remain solvent". Several of those who made a ton of money nearly went bankrupt holding onto their bets for just the 2 years they knew it would take for them to pay off - and they were right about the 2 year prediction!


I'm still confused by the CDS bit. Is there a good eli5 article on it? I've already checked the eli5 subreddit and couldn't find anything specifically on the mechanism that allowed shorting.


Sure.

The short answer is.....

You(party A) and party B have a contract.

You want to eliminate the risk that Party B goes bankrupt and can no longer pay you.

You go to party C who will take small payments from you each month/quarter and in return they will pay you if party B goes bankrupt.

Now the important things to note are:

1) The amount you pay party C is based on how you and party C analyze the bankruptcy risk of party B.

2) The amount that party C pays you when the CDS is triggered doesn't necessarily have to have any relation to your original contract with party B. ie party B might be making a one time payment of $1 million to you but nothing stops you from taking out $10 million in CDS protection on party B.

3) The trigger for a CDS may not necessarily be the failure of party B to pay, ikt might be related to their credit rating, ownership changes, or really any term you can negotiate.

4) you can write a CDS when you aren't even involved in the original party A to party B transaction. This is what happened in teh big short where Michael Burry had no involvement in the Mortgage back securities, but had several banks writhe him a CDS that payed out if the original mortgage backed security had a certain percentage of its mortgages fail.

This is where things really went off the rails as companies like AIG wrote the CDS protection for a whole lot more MBS(Mortgage backed securities) than they could possibly ever cover, because their models predicted that there really wasn't any way they could fail in the manner they did. Suddenly a single MBS could have 100x its value in CDS default protection written on it. This leverage is how the 2008 bailout got so big.


You're missing the very best part ... in many cases, the C parties allowed customers to assemble their own CDS specifically containing the very worst segments of the contracts between parties A and B (to stretch your analogy).


It's not an analogy - you are confusing the difference between CDS (which is what chollida1 described) and MBS (a type of collateralized debt obligation the is pooled together and various buckets of risk are tranched and securitized, which is the aspect you describe).


The 100x is more because companies weren't really able to sell these contracts. Instead, they'd make another one with the opposite position to mitigate their risk. In case of default, there'd be a lot of offsetting payments.


I think the comment is referring to the fact that one can write CDS on top of an underlying instrument. So, if I have $100 million in MBS, other parties, or even myself, can write CDS contracts that reference my $100 million MBS.

What I have just done is to create more exposure than the underlying MBS. The MBS is still only $100 million, but the total outstanding contracts on the pool (MBS + CDS) is now much larger.

What you are talking about is related to offsetting contracts, which is different.

AIG was left holding the bag for most of these. They are one of the few who did not offset like you are saying.


> They are one of the few who did not offset like you are saying.

Which just goes to show important the offsets actually are in practice.


Why is that type of betting (gambling) allowed? That's just plain insanity.


It's not betting (gambling), it's insurance. Is insurance insane?


How could you call it insurance when you have no interest in the original security? What are you insurance against?


@jkimmel is 100% right: You don't need to be a direct party in a contract to be affected by it. Here's another example:

I buy a house at a higher-than-usual price for the area, expecting its value to increase because of some development that is occurring nearby—an automotive manufacturer has recently agreed to build a plant about 10 minutes away, creating around 2000 jobs in the area.

However, the deal still needs to be approved by the Feds. If I wait until the approval, the house might cost even more to buy and potentially make my investment unprofitable (or not profitable enough). But if the plant is not approved, my house will likely not gain enough in value and I will have paid too much.

To insure against the Feds not approving the plant, I can by a CDS against that risk. I will lose a small amount of profit in my eventual sale of the house to pay for the CDS, but will also prevent a large (profit) loss in the event that the plant is not approved.

Note: In this example, I do not have any direct interest in the (future) plant, or the auto manufacturer, and I obviously do not control the actions of Federal regulators. Nevertheless, I am exposed to risks (and benefits!) taken by them as a homeowner in the area, and a CDS can be used to insure against those risks at minimal cost.

Admittedly, this is fairly sophisticated for an individual homeowner, but this kind of thing is routine for financial professionals. CDS is a kind of insurance, and it's totally legitimate, even though you are usually not insuring something you have a direct interest in.


I see. Thanks for the clarification. I would definitely still consider that insurance -- as you said, you're really insuring your vested interest in the property. If you were doing such and you didn't even have an indirect interest, I would consider that gambling.

Of course, this is just my own personal subjective consideration. Not related to the official definitions of these terms at all.


You don't need to be a direct party in a contract to be affected by it.

You can insure against possible negative outcomes of someone else's contracts.

e.g.) Party A and Party B sign a contract that Party B will pay Party A $X by some date. Party A has seperately agreed to pay you $0.9X shortly afterward.

You are a savvy businessperson and realize that Party A will not have $0.9X to pay you if the Party B fails to make their payment.

Party B is on the rocks after a nasty reorg, and you think the probability they fail to pay is significantly greater than 0.

You want to protect your business against this event. To do so, you purchase a Credit Default Swap from Party C for $0.05X that ensures the full $0.9X payment.

While Michael Barry was "speculating" using CDS's, you could also argue that he was "insuring" his firm against a housing crash, which did tank most investment vehicles.


Insurers (eg casualty insurance) have no interest in the underlying. They are pure speculators. Gotta have a counterparty.


As I understood it in the Big Short:

Christian Bale's character's hedge fund wanted to make a big bet that mortgage backed securities would fail.

So he shopped the banks for those that would take that bet.

The ones that agreed to take the bet, he paid them a negotiated monthly amount (basically rent) to keep the bet in effect. If he stopped paying rent, the bet was off.

The banks thought it would never happen and it was free money coming in. Since they thought the risk was 0% it was free money coming in.

... basically, the banks thought he was crazy, but would take his money to keep the bet going.

In the movie, Bale's fund gradually loses money paying "rent" on this bet / option / contract. Things get dicey when conditions that should result in the bet paying off get covered up by the banks and government, and the investors in his fund get pissed they aren't getting any return on their money as promised.


It wasn't a bet - it was insurance, and his payment was the premium to underwrite that insurance. Of course the banks who wrote him the CDS failed to price the risk appropriately...

It's like you are arguing that life insurance is a bet.


It's just insurance. One side pays the other side if a certain outcome occurs. In exchange for protection from the outcome, the side seeking protection pays a fee.


It's more like buying insurance on someone else's house, when you know they are a meth lab and the insurer doesn't.


In that case the insurer should do their due diligence before taking the bet.


A hand wavy analogy is that a CDS is bankruptcy insurance.

So the guy in the movie got the bank to write him an insurance policy on the mortgages (really mortgage derivatives) that other people held. When those mortgages failed, the policy paid out.


I forgot the eli5 part.

Let's say you have a little sister. It's been a little while since she screamed about her stuffed frog, but you think she is going to do it again soon. You come to me and say that I can have one of the cookies out of your snack every day if I agree to give you a whole bag of cookies next time your sister screams about the frog.


Pretty much everyone is saying that interest rates are going up in the US and are going to go up soon. Alan Greenspan just came out today saying people are going to be surprised how quickly they go up.


> Pretty much everyone is saying that interest rates are going up in the US and are going to go up soon.

I wouldn't say that.

According to Bloomberg's interest rate predictor the probability of a hike before the end of the year, ie the definition of soon, is less than 50%.

Even if you go out to the end of 2017 you still only get to 67% probability of a fed hike. And 67% is a far cry from "pretty much everyone".

And even with all of that, the most predicted rate hike is a minuscule 0.25% to 0.5%.

Now, I agree that this is all up for debate and markets can definitely get this wrong, but I stand by my assertion that low rates are here to stay. I really don't think the years of 5-8% rates are coming back any time soon.

But I'm always willing to change my mind if someone can convince me, so back to you.....What makes you think that this era of low rates is going to end?


Not that people's bets mean much, but you can look at how financial markets expect rates to move using the interest rate futures market. These contracts allow you to infer what the market really expects rates to do (as people are putting their money where their mouth is - to the tune of billions of dollars rather than an off the cuff comment on a financial news show). Free data for the 30 day future contract/bet is available from Quandl - which is a great site for checking out free financial data: https://www.quandl.com/collections/futures/cme-30-day-fed-fu...


Or just look at government bonds. Investors wouldn't be investing in 10-year bonds with negative yields if they thought interest rates were going to go up significantly.

I don't work in the financial industry, but just as a bystander it seems clear to me that there are a lot of bond traders, like Bill Gross, who have strong interests in rates going up significantly. Just like small-cap, pump-and-dump scammers, they're trying to move markets with their sophist predictions.

I don't think there's much of a dispute that there'll be a regression to some mean. But judging by where people are actually putting their money, the time horizon appears to be much further away than many of the pundits would have us believe.

Also, Japan....


Interesting, I hadn't read the outcome of Wednesday's meeting. Prior to it the local Fed Chiefs were saying that they definitely expected one to two raises this year, and maybe as early as September. All rate hikes are usually minuscule, but they add up. 2 minuscule rate hikes would double the rate right now.

One reason to raise rates is that they were set as low as they were for emergency purposes. The emergency itself is over, so time to start increasing them a little.


We've been hearing about rate hikes Real Soon Now for more than a year now. Have they actually happened? I haven't heard of any.


There was one in December. They paused because Brexit actually happened.


I remember listening to an EconTalk podcast (highly recommended!) after the quantitative easing started. Very interesting.

Typically lowering interest rates pushes inflation higher. When your economy is sluggish, that's a good thing, but once it picks up steam, it can cause runaway inflation.

The challenge raised is that there is a delay in the response of the economy. Usually interest rates aren't increased dramatically until inflation is already underway. Otherwise, if you raise rates when the economy is only doing "a little" better, you risk squashing the growth.

So what happens is that inflation starts to go up and by the time the Fed has the balls to really raise rates (because it's plainly obvious the economy is doing well), it's too late and you really overshoot your inflation target which causes rates to rise dramatically.

Overall, I would agree that it seems unlikely interest rates will go up in the near future. However, that can change really quickly.

Back in 1979, the interest rate when up 400 basis points from 10% to 14% in less than a year.


nice definition of 'soon' you pulled out of your ass.


Why does Alan Greenspan's opinion have any weight? He is not in any kind of position to have control over rates.


He should know: he's probably the individual with the single largest share of responsibility for the housing crisis.

Not to mention responsibility for setting up with swindle that is currently used as an excuse by those who want to feed Social Security to the ibankers.


No, but since he worked in that position for years, he might have some kind of insight?


Insight on how to be incredibly wrong on a housing bubble?


Easy to say in retrospect. Wrong like nearly everyone else.


Saying "nearly everyone" was wrong and blindsided by it is a bit of a stretch. Not everyone that saw it coming profited from it big like the now fabled protagonists in The Big Short. But to people who chose to pay attention to it, the issue was known and well publicized ahead of time:

https://youtu.be/tZaHNeNgrcI

https://youtu.be/kFd8YluIVG4

The key to 'seeing it coming' of course, is to first willingly accept the possibility that something might happen, and then actually look into it. The issue was that many people were just not entertaining the possibility at all, and therefore not bothering to look into it. The information was all there for anybody to see, but this was a case of people missing the gorilla in the room because they were too busy paying attention to the game[0].

[0] http://www.livescience.com/6727-invisible-gorilla-test-shows...


Nobody with any experience was surprised by the housing collapse. The only people I know who were surprised (in the financial industry and out) were people who've honestly never considered the numbers for even a minute.

Housing prices have to track rent, to some degree, and vice-versa. One of them will change to match the other and wages were not going up...

All the ARMs and layers of CDOs are irrelevant. They multiplied the results but didn't change the underlying issue. Leverage doesn't help you when the fulcrum shifts.


Or you could say, nearly everyone else was wrong as a result of him being wrong.


I don't know if interest rates will go up or not, but when rates do go up, they can be swift.

The Adjustable Rate Mortgages offered by for example First Republic Bank are referenced to one-month LIBOR, and it went from 1.11% in May 2004 to 5.11% in May 2006, i.e. 4%-points in 2 years.

On a $1M mortgage, that's $40k per year in swing over two years (typically deductible, so you'd have to factor in your marginal tax rate to get effective cost).

http://www.macrotrends.net/1433/historical-libor-rates-chart


It's important to mention that they've been saying that for 7 years now, and that Alan Greenspan has been wrong about everything.

We haven't even been able to get inflation up past 1.7%.


Agree on Greenspan, but do you really believe that the inflation rate has been 1.7%? Actual inflation is estimated to be anywhere from 3-7%, depending on the source.

The Fed, setting interest rates way below the actual inflation, is actually taking money from savers and retirees (rather than rewarding them), or forcing them to be reckless with their money by investing in principal-at-risk investments (which they should not at their age)


What source is that? The ones that don't say inflation is 1.7%?

I'd like to point out as the other poster did that you shouldn't look at housing in development constrained areas (though that's a very interesting sidebar to the whole thing - how much of development constraint exists to prop up housing prices)?

Incidentally "official" inflation in the SF Area is 2.7%, still not 3-7%. (http://www.bls.gov/regions/west/news-release/consumerpricein...)


If you live in a growing but development-constricted city, housing inflation alone would bring your personal inflation rate above 3%, but besides that (e.g. if you live in Houston) the prevailing inflation rate is still low.


I think everyone would like for them to go up, but I haven't seen any real reason why it might happen. The rich world has been stuck in very low interest rates for a long time.

(Edit since I'm fixing a typo: Look at the treasury yield curve. Has it gotten steeper, which would indicate an expectation of rising rates? No. The 30 year rate has been falling the last few years.)


Here's a link to the Greenspan interview: http://www.bloomberg.com/news/audio/2016-08-18/a-closer-look...


>And the most accurate part of the movie was that it showed the anguish that each of these people went through. They'd seen something no one else had, been right about it, had the event happen and even then everyone else continued to plug their ears and pretend nothing was going on while they continued to bleed money each month. It really did a good job of showing that you can be right about everything and still not make money if the rest of the market continues to act insane.

You mean: it showed the part where the assholes had grabbed control of the government and rigged the market, pointing a gun to the head of the world economy, so that their idiotic bets would pay out by force, thus penalizing everyone who had bet on the facts, right?


>I'd agree that the US housing and derivatives markets are in a much better position than in 2007-2008

Could they have been worse?


The fact that we were not able to ban mortgage securitization after the crisis is appalling. IMO this is the root of the problems. Make banks hold their loans on their balance sheets!

I also think the point about banks having capital buffers is laughable and not logical. It's like claiming we've solved chronic migraines because we have a stash of Tylenol. The cause has not been eliminated. And let's say there's a crisis and that bank capital is used to cover the losses. Now what? You now have banks which according to our own definition are under-capitalized. They used all their capital for yet another crisis. So now your banking system is weak again. Who is going to give money for these banks to capitalize again if they keep having crisis after crisis? Ah that's right, the taxpayer!

A nation's banking system reflects its cultural values. This is quite true about the USA.


>The fact that we were not able to ban mortgage securitization after the crisis is appalling. IMO this is the root of the problems. Make banks hold their loans on their balance sheets!

Absolutely. The problem is that mortgages have become speculative instruments, with the profit privatized and the risk socialized. The old system was far more stable.


The old system gave us the savings and loan crisis. Is it fair to say the alternative is to recognize that for many homeownership simply is not an option?


It sounds like a reasonable theory on its face. However, the outcome of this is likely that banks will stop making loans all together. If they were to offer mortgages at all, it certainly would not be a 30 year fixed rate product.

It's not correct to say that banks can sell off all of their risk through securitization. They do in fact retain some of the risk. Though most has been transferred.

The buffer is (in theory) in excess of the losses. So, after a crises a bank would be left with the buffer and be able to continue to operate as a going-concern. The question is how to set the buffer and how big ?


>However, the outcome of this is likely that banks will stop making loans all together.

Really? We had a healthy mortgage market for a great many decades before mortgage backed securities were invented. And if mortgages are such a bad idea that banks can't make money off of them, then we shouldn't have them.


The secondary mortgage market isn't a recent invention. It's been an explicit policy goal of the US government at least since the Great Depression.


It's quite simple to get a mortgage in other countries where banks don't securitize. And if you get into trouble they will work with you rather than immediately foreclose.

Fixed rate mortgages for 30 years are a scam. How can a consumer predict interest rates over 30 years? If they go down they have to spend money on fees to refinance which benefits the Lenders. Adjustable rate mortgages are far better and allow for quick stimulation to consumers when rates are cut.


The complication here is that local banks do not always have sufficient capital to finance all the mortgages in a growing community.


Partly because the state charges too little for the guarantees it offers, taxpayers are subsidising housing borrowers to the tune of up to $150 billion a year, or 1% of GDP.

The original post makes this claim without any citation. I don't think it is correct. Recent stress-test reports on Fannie and Freddie show that, even making provisions for future bailouts in the stressed scenario, the agencies are profitable to the government. [1] Instead of costing $150 billion per year, since the agencies were nationalised they have returned $60 billion to the government.

[1] https://www.bloomberg.com/view/articles/2016-08-08/fannie-an...


The post links to another article with a few more details:

[Imagine that the securitisation industry] had to carry the same level of capital as banks do and to make an adequate (10%) post-tax profit on that capital. The higher costs entailed give a sense of the scale of the current distortion. On this basis The Economist calculates the subsidy on mortgages to be running at $150 billion a year, 1% of GDP. (This estimate includes the impact of the Fed’s bond-buying on interest rates and the cost of tax breaks on mortgage-interest payments.)


This stress test is assuming credit losses about a quarter of the size of the stress test 2 years ago, and much smaller than the last crisis (0.56% vs. 2% vs. 1.4%). Does anybody know if something has structurally improved in the Fannie/Freddie portfolios, or whether the testers are just less pessimistic than they were 2 years ago?

[1] http://www.fhfa.gov/AboutUs/Reports/ReportDocuments/2016_DFA...

[2] http://www.fhfa.gov/AboutUs/Reports/ReportDocuments/GSEFinPr...

[3] https://online.wsj.com/public/resources/documents/irasohnfin...


If anything this [1] article would lead the reader to believe since Fannie/Freddie are now spearheading the leading charge their portfolios are further out of line with the market reality.

[1] http://www.economist.com/news/leaders/21705317-americas-hous...


The claim of the article is more that current government policy has not correctly priced in the externalities caused by most mortgage funds ultimately being available through bonds rather than through bank deposits. "The public would have to foot the bill, of around $400 billion, making explicit the contingent liability for future losses that it already bears. The cost of mortgages, at a record low today, would also rise. But that would eliminate the ongoing hidden subsidy and create a level playing field so that private firms were able to do more mortgage lending. If that bill was too big to swallow, a second-best would be to impose the new rules on new mortgages, leaving the stock of subsidised existing loans to run down over the coming decades." Thanks, by the way, for linking to the Bloomberg article, about a different issue, but a good source in general for comments about economic policy.


I can remember people making similar arguments before the last crisis. It turned out we had to put not only bailout Fannie and Freddie, HUD's mortgage programs and the entire mortgage banking industry but also create a price support program for mortgage securities.

If the guarantees really cost nothing or are profitable, the US should stop making them. They create all kinds of distortions even if it averages out.

But it's the fact that all the particpants (banks, agencies, etc) want them so badly makes it clear how valuable they really are.


It's possible to turn a profit while still selling something too cheaply. The claim (as I understand it) is that leaving money on the table is tantamount to giving it away.


To sum it up, despite there being regulations on CDOs the real problem still exists. That problem is that loans are easy to get the gov stamp. Thos makes it easy for new home owners to get loans. And this makes it easy for sales people to sell the loans.

Much of this opinion article is based on an other article [0] which seems less doom and gloom and more about the shift to a majority of state sponsored loans. This in itself its own issue in that capitalism isnt doing what it should be when everything is subsidized by the gov.

Ultimately these two articles seem to be about responsibility. Banks are not responsible for homeowners making poor decisions. Banks are not responsible for investors trusting a market that is not self sufficient. And Banks are not responsible for the government to enable them with tax payer money. Meanwhile everyone can point the finger back to the banks and the bank depends on the fact that too many economies rely on them to fall

[0] http://www.economist.com/news/briefing/21705316-how-america-...


This in itself its own issue in that capitalism isn't doing what it should be when everything is subsidized by the gov.

Thus $500k 800 sq ft bungalows. You show me a chronic bubble and I'll show you a 'market' riven with subsidies, tax incentives, government secured debt and pencil whipping regulators. Healthcare, education, housing.... Yet when the bubble pops and the too-big-to-fails have to be 'recapitalized' all we get from the ruling class is more hating on markets.

Your last sentence is interesting:

the fact that too many economies rely on them to fall

I believe you meant that banks can't be permitted to fail. But it could also be read to mean that what we've self inflicted is a system whereby the greater fools are those found holding worthless CDOs (and all of their various derivatives) when the bubble pops; the market is ultimately expressing itself through the failure of financial institutions.


"Thus 500k 800 sq ft bungalows"

This may sound cynical but that is one of the fair markets that likely will not fall. These are luxery that are likely only to be bought and sold to people looking for vacation spots or a crashpad. Much more likely to be paid off than a 4 bedroom in the midwest bought by a hardworking train engineer and his wife working part time in a restaurant.

"We've self influcted a system wherevy the greater fools are those found holding worthless CDOs"

I really appreciate the hot potato analogy! Exactly what I was thinking.


> 500k 800 sq ft bungalows

For what it's worth, this describes a significant portion of the single-family housing stock in Seattle.


>Yet when the bubble pops and the too-big-to-fails have to be 'recapitalized' all we get from the ruling class is more hating on markets.

The reason this happens is that during the boom phase, the conservatives and libertarians, the defenders of the market, tell everyone things are going just fine. And then when it all blows up, they suddenly change their line and say it was all due to government subsidies and regulations. That is what happened last time, remember?


Why does the government need to be involved in mortgage lending at all? Is there a specific market failure inherent in it people can point to that calls for government intervention to overcome (akin to the public good failure for e.g. roads)?


People who have nothing to lose tend to make poor economic decisions and poor social decisions.

That's why home ownership matters. It's a poor investment vehicle only if you assume people would invest savings from renting into appreciating assets, rather than burn it on consumables. They won't.

So, yeah, government is involved because it's a collective action problem. Individuals and the market left to their own devices will reach a suboptimal equilibrium.

Is home ownership the _only_ answer? No. The problem is fundamentally cultural. But in the U.S. especially, subsidizing home ownership seems to be the most viable mechanism we can, as a society, agree upon. Probably because of our emphasis on individualism. In a society that highly values individualism, people will be more incentivized to strive toward home ownership than other alternatives. We went them (us) to delay other, less economically efficient forms of gratification.

In some Asian societies, for example, different cultural preferences make alternatives more viable. For example, they're more likely to emphasize investment in education. I'm pretty sure the phrase "book learning" isn't derogatory in most Asian societies, even for low-skilled laborers who feel disenfranchised and lack any intellectual ambition. (Not that those things are necessarily related--I strung them together for emphasis.)

In any event, the point is to keep people invested in the economy in such as way that their individual choices tend toward more productive activities. It's a channeling function.


The government, acting as the collective will of the people, believes, due to recent experience, that a housing crash, and the concomitant cascade of mortgage defaults, is something worth making an effort to avoid?

I think that's how the story is supposed to go.

I guess the flip side is that quote by that fellow I read somewhere that went something like: democracy cannot last as a permanent form of government, it can last only until people realise they can vote themselves largess from the public purse.

Let's wait and see what happens.


Homeownership was a major pillar of post-WW2 cultural stability among the middle class, so subsidizing it was considered a social good.


Cringed while reading. We've got to stop transcribing intuitions into universal laws if we want to claim the sapiens in homo sapiens.

Regardless of political stance, for any animal to require half its lifetime in effort to acquire a shelter, well that is retarded. And for some reason we consider ourselves more enlightened than our ancestors building functionally useless monoliths. At least the pyramids united the tribe or whatever.


I feel like I can relate to your position however I believe there is much more to this than simply "unevolved"

- a hive mind (similar to that of bees or ants) likely considers death to be a reasonable loss to the point of canabalism being deemed reasonable. This is probably the closest to full centralization there is.

- in gorillas there tends to be one male per multiple females and per area. There also tends to be a female pecking order where one stands dominant and weaker ones get scars. This is probably the closest to fully independent since each individual only gets what they take.

We live in a world where money can be anywhere and we want to allow anyone to have the opportunity to get what they want (so long as it doesnt harm the collective good). What tends to happen is the money becomes centralized to trade centers or with rare resources like peaceful community or opportunity in entertainment. However, space is a scarce commodity in these locations because it attracts so many people. So isnt it fair to increase the prices? The people who are making the most money for the area should have the most convinient location.

"unevolved" implies that there isnt rhyme or reason to this.


When these institutions were first created no one imagined the harm they would inevitably cause. Now that they exist they can't be killed, because they help some people. Like a lot of things, really.


Thats not entirely true. England has had a number of economic scandals before the US recieved their independence. There is also certainly a president of "too big to fail" to the point that the king was invested [0].

[0] https://en.m.wikipedia.org/wiki/South_Sea_Company


For an entertaining summary of this, I recommend the youtube series Extra History: https://www.youtube.com/watch?v=k1kndKWJKB8


Haha I thought we were talking about Fannie Mae, not USA itself. Although, the latter might be a better comparison to the South Sea Company...


One specific market failure people can point at is government involvement.


That problem is that loans are easy to get the gov stamp.

It's more that the fee the government charges for its stamp is too low, such that if the loans sour, the taxpayer ends up paying back the loan.

Partly because the state charges too little for the guarantees it offers, taxpayers are subsidising housing borrowers to the tune of up to $150 billion a year, or 1% of GDP. Since the government mortgage machine need not make a profit or have safety buffers, well-run private firms cannot compete, so many banks have withdrawn from making mortgages. If there is another crisis the taxpayer will still have to foot the bill, which could be 2-4% of GDP, not far off the cost of the 2008-09 bank bail-out.


The simplest approach would be ... The nationalised mortgage firms that guarantee the bonds ... should be forced to raise their capital buffers and increase their fees

Raising capital requirements seems like a minor tweak, a band-aid cure.

I would argue that the systemic risk could better be managed by changing the fundamental economics of the financial instrument in question: the classic residential mortgage.

I'm surprised the Economist didn't mention the short-selling idea which Richard Schiller has been pitching [1] since the Financial Crisis.

The basic idea is to make it possible to accurately short-sell residential real estate. This would allow homeowners and lenders to be partially protected against declines in real estate prices (the risk being transferred to third-party speculators and investors). Just like a lender requires homeowners to buy property insurance, it could also require homeowners to enter into financial protection against declines in property value. This would protect the homeowner against negative equity (where the property value falls significantly below the borrowed principal). It was because of negative equity that so many homeowners defaulted on their mortgages during the Financial Crisis, leading to a vicious cycle of foreclosures, sales and declining property values.

[1] http://www.nytimes.com/2015/07/26/upshot/the-housing-market-...


the "protection" against negative equity is the down payment. negative equity only matters if incomes take a hit and sales are forced.


The down payment is equity, it's not really the same as protection. Although the higher your down payment, the lower your leverage risk. Let's say you put $100k down to purchase a property for $500k (i.e. you borrow $400k). If the property value falls to $400k (20% drop), your entire equity is wiped out. If you sell the property, the proceeds go to the bank to pay back the loan, and you have lost 100% of your the original $100k down payment ... This is due to leverage. If the property value falls below $400k, then you are in a negative equity situation ... Get deep enough into negative equity and it becomes rational for some people to just default and "walk away."


very few non-recourse loans in the housing market these days. a 20% decline in housing prices, and there's limited - zero systemic risk (ie the banks via linked derivatives that the author's so worried about); only the owner loses.


The linked article seems to be a bombastic summary of a longer, more neutral, and better-supported article [1]. Strangely enough, they were published on the same day and apparently are both featured in the print edition? It seems like The Economist is doing A/B testing to see what draws more outrage from its readers: "Nightmare on Main Street" or "Comradely Capitalism"?

The linked article is confusing and uninformative. I suggest reading the longer version.

[1] http://www.economist.com/news/briefing/21705316-how-america-...


The linked article is an editorial, the other article the news article. The Economist calls their editorials "leaders," which I agree is confusing. But its standard practice to publish both an editorial and a news/information piece in the same issue.


One other thing to keep in mind is that there was a great boom in building in the US during the 1970s and 1980s, and many of the homes built during that period were built using substandard building materials and techniques, many of which are no longer up to code for new construction but were grandfathered in.

How many of those millions of homes are actually solid, reliable structures? How many suffer from moisture damage, mold, or have been poorly maintained? What kind of cement was used to pour the foundation? What kind of insulation? How many insect or rodent infestations has the house suffered?

There is a vast difference in quality between an old building in a city that has been renovated for apartments compared to a suburban home that has essentially gone unmaintained for decades and was built as part of a speculative housing development project by someone who had no long term interest in the quality of the construction.

I think we'll start to see a gradual crumbling/disintegration of suburbia as the houses fall into disrepair and the mortgage debtors are in too far over their heads to afford proper maintenance.

The right thing to do in 2008 would have been to let the prices fall and stay down, to encourage proper renovation and maintenance.


Do you have any articles/books detailing construction practices during that time period? Also any numbers on what percent of housing in America was built during that time period.


> I think we'll start to see a gradual crumbling/disintegration of suburbia as the houses fall into disrepair and the mortgage debtors are in too far over their heads to afford proper maintenance.

This hits the nail on the head.


Can anyone provide any background/documentation on the claim "Now 65-80% of new mortgages are stamped with a guarantee from Uncle Sam that protects investors from the risk that homeowners default"? I don't really doubt that it's true, I'd just like to learn more about what that is and how it works.


Meaning, the mortgages (default risk) are guaranteed by the government, either explicitly (FHA) or implicitly through a government sponsored interprise (Fannie and Freddie).

After the housing crisis the private market (mostly banks) left very quickly. FHA and the GSEs stepped in and took their share.


Does the FHA issue bonds? (Turns out this stuff is nigh impossible to Google.)


Yes. FHA and the GSEs pool mortgages from banks, stamp them with their guarantee, and then sell the loan pools as MBS to investors.

(Most) banks, especially small ones, are unwilling to sell and hold a fixed rate 30 year mortgage. There's way too much risk, price volatility (driven by rates) and uncertainty about what will happen over a 30 year time period.

Banks only do a 30 year mortgage because they know they can sell it to FHA or the GSEs.


The last crisis was also heavily related to sub-prime mortgages. How much has sub-prime lending been mitigated?

Before the last crash, you could get "no paperwork" mortgages, you'd just to have a little higher interest rate. Lenders would also squeeze people into ARMs so their initial rates were really low and then would go up after the adjustment period. And nobody cared about things like PMI because prices were going up so quickly. You'd have enough equity in your house if a few years and the PMI would go away.

The general consensus was that if you were in the market to buy, you should buy as soon as possible before prices went up much more.

Thankfully, I live in an area that didn't see the enormous run up of house values. The crash affected our housing and we still had the normal negative effects of a big recession. Housing prices have certainly recovered by now and I do wonder how much of it is still smoke and mirrors. But in markets like Florida and Arizona where house prices got really out of control, have those areas also recovered? That would be more shocking to me.


As far as I know, the last crisis wasn't heavily related to sub-prime mortgages. The primary predictors of default were the amount of money down and the numeracy of the borrower, not the borrower's credit score.

Plenty of people with excellent credit scores spent too much money on a house and ended up defaulting, and even people who could afford the payments ended up severely underwater; houses became very expensive very quickly, you couldn't buy one without spending too much.

Blaming subprime is largely a right-wing narrative. The worst thing that happened in subprime is that a lot of people (especially minorities) who could have qualified for mortgages on better terms were steered into subprime and ARMs because of their worth in high-growth (and ostensibly low-risk) CDOs.


I'm really curious about the things you are saying since they are so different than what people regularly say on the subject. Especially the right-wing narrative part. Can you suggest any sources on all this?


> The last crisis was also heavily related to sub-prime mortgages. How much has sub-prime lending been mitigated?

http://www.qualifiedhomeloans.com/campaigns/statedincomeloan...

http://www.forbes.com/2010/07/02/return-liar-loans-personal-...

No. Its still essentially the same with no paperwork mortgages with higher interest rates and ARMs. Very little changed other than a few regulations that might stop a systemic collapse of large banks.

> Thankfully, I live in an area that didn't see the enormous run up of house values. The crash affected our housing and we still had the normal negative effects of a big recession. Housing prices have certainly recovered by now and I do wonder how much of it is still smoke and mirrors. But in markets like Florida and Arizona where house prices got really out of control, have those areas also recovered? That would be more shocking to me.

http://www.tradingeconomics.com/united-states/case-shiller-h...

The market as a whole is almost back to where it was in 2008.


You can't read that Forbes link and say "no" if you know a little bit about finance:

> Wall Street Funding of America [...] circulating offers to make low-doc loans to borrowers with credit scores as low as 660 FICO, as long as the borrower was self-employed, seeking no more than 60% of the value of a home and had six months of mortgage payments in reserve.

660 FICO is not sub-prime by any standard, and it requires 40% down AND substantial reserves. That risk profile is great, and 180 degrees away from the old stated-income loans.

Anecdata: I've opened 4 mortgages since the crash, 1 with a big bank and 3 with smaller lenders. Every one of them required substantial and verified documentation, and my FICO is just under 800.

Lending is nowhere close to the craziness it used to be!


> 660 FICO is not sub-prime by any standard

Experian buckets 740-830 as "super prime" and 680-739 as "prime". I'm not aware of any credit agency that buckets 660 as "prime" (or better).

Some mortgage originators consider 670-690 and no 30-day lates in the last year as "near prime".

660 is indeed "sub-prime", IMO.


You do realize the other link is a stated income / no doc loan right?

http://www.bloomberg.com/news/articles/2015-01-28/get-ready-...

> Angel Oak is willing to buy loans with credit scores as low as 500, though the average has been about 670, just below the 680 level some use as a cut-off for subprime on a scale of 300 to 850, Hsu said. The company makes sure borrowers can afford the payments and requires at least 20 percent down payments, he said.

660 is subprime.


That is not a no doc loan. You conveniently left out the very next sentence:

> The company makes sure borrowers can afford the payments and requires at least 20 percent down payments, he said.

"makes sure borrowers can afford the payments" is code for "documentation required". What's more, 20% down gives good capitalization and is hardly sub-prime (as a loan product).

Low-score low-down FHA loans have existed for a long time and are not a cause for concern -- mortgage insurance is required and offsets the risk, unlike the crazy balloon ARM products of the bubble.


http://www.qualifiedhomeloans.com/campaigns/statedincomeloan...

Did you really not read the first link I gave you which is a stated income loan?

You are very interested in convincing me you are right when I know these no doc places exist and know people who have loans with them.

I was simply correcting your belief that 660 was not "subprime".


Dude, come on. Embedded youtube video that doesn't exist anymore, 2000s-era site -- you really had to scrape the bottom of the barrel to find that page.

See my sibling comment, most lenders don't consider 660 subprime.


> Dude, come on. Embedded youtube video that doesn't exist anymore, 2000s-era site -- you really had to scrape the bottom of the barrel to find that page.

It was an ad on the same google search as that first article.

Look, these products exist. You can live in denial if you'd like.


For those confused on what sub-prime fico scores are:

"Historically, subprime borrowers were defined as having FICO scores below 640" -- https://en.wikipedia.org/wiki/Subprime_lending#cite_note-Lo-...

"550-620: Subprime." -- http://www.credit.org/what-is-a-good-credit-score-infographi...


NIMBYism is already putting places like Palo Alto, Boulder, etc... well out of reach of the middle class, with a class of "haves" that live there, and "have nots" who have to drive in to work. What would happen to this situation if you killed the subsidy off?

What would it do to the dynamic between generations: older people got a big leg up, and you get nothing?


It would cause alot of anger. I hear a lot of hatred at work towards the baby boomers about everything


Niall Ferguson called this several years ago.

http://www.telegraph.co.uk/finance/financialcrisis/9338997/R...

His Reith Lectures on the subject (Burke's social contract and its implications today) are worth putting into your mp3 player.

Ctrl-F niall

http://www.bbc.co.uk/programmes/b00729d9/episodes/downloads


Thanks didn't know about that, and podcast link is super great!


House prices are like bond prices. When interest rates are low, prices go up. When rates rise, prices will fall. People buy houses based on the monthly payment they can qualify for. Right now, prices are high because the low interest rates mean lowish monthly payments on inflated prices. But when rates rise, the value of the houses will fall because buyers will not be able to qualify at the previous inflated prices. This will mean that a lot of homeowners will be underwater again. Even though they should be able to afford their payment, there will be a similar level of panic.


In retrospect, I wonder what would have happened if we let AIG and the banks fail. The FDIC would convert them into DINBs with the insurance limit raised to $250k per account holder. All debts including mortgages would be discharged tax-free automatically (making the topic of mortgage notes moot). Bank charter restrictions would be loosened so that new banks can be created as soon as possible.


Whatever the implications of letting the debt just "fall on floor" would have been, it's worth noting that the decision to not allow that outcome was reached years before the events themselves. Ben Bernanke's helicopter money speech was at least one demonstration of this.


As I was driving into the office today, passing all these auto dealerships, big, small, and mom and pops small. I wondered if they have an exit plan. In 5 years or less, every automobile manufacturer will have a self driving car. Will it make their inventory obsolete or a lot less valuable? Will it make my car a lot less valuable? On a side note: http://www.visualcapitalist.com/chart-largest-companies-mark...


Has anyone heard of this business? Why did I not know about this company around 2010 when I needed a loan modification?!? They are in 42 cities in the U.S. and have been around for 22 years. No down payment, no closing costs, no fees, no income limit, no need for perfect credit! https://www.naca.com


>Partly because the state charges too little for the guarantees it offers, taxpayers are subsidising housing borrowers to the tune of up to $150 billion a year, or 1% of GDP.

The current environment offers housing borrowers money at a price below the market.

The simplest bet with this cash (buying a house and holding) has probably been overdone in the market. House prices are quite high now.

Still, small but crafty borrowers should be able to profit with this cheap money.


Buy and hold is fine, but that's a strategy like buying Blue Chip stocks. Early 2000s real estate was like the options market -- super short term.

If you can hold a house 10 years, chances are very, very good you won't be underwater at the end.


Are you suggesting that small crafty borrowers spend the money on something besides the mortgaged house? I think that would be fraud. Or maybe you're just saying they should sell the house quickly? How would that profit?


Yes it has not been fixed, but it's a supply-side problem and not a demand-side problem. More houses should be built to meet demand, but this is not happening in many cities.


I am still of the opinion that the government's maximums for its lending are too high at 417k to 1.8m depending where you live.


You should be more concerned about downpayment amounts (3% is acceptable, and some lenders will finance with 0%). This is what pushes home prices up artificially.

I had a foreclosure where I stopped making payments in 2011. My foreclosure finally completed in December of last year (4.5 years later). I already have lenders who will finance me on a new mortgage.

It's madness.


Yeah. Was speaking to a realtor yesterday actually about this (just an acquaintance) and homes in some areas in a particular midwest city have risen in value 23% in one year. As soon as a place goes on the market it has 10 offers over the asking price. This isn't good. But if you have a great job and good job security it is idiotic to not buy a house right now with the low interest rates assuming it's not way overpriced (and these can be found).

I purchased a home that costs about twice my annual salary. I think it's reasonable. I put 0 down and walked away with a check at closing after seller closing costs for some items I wanted repaired.

Anyway this is just my general commentary and my experience. I don't know enough about the market as a whole to make any real informed contribution, but just wanted to share anyway.


Where are you getting those numbers?

Looking at https://entp.hud.gov/idapp/html/hicostlook.cfm

I see the majority of FHA limits in Texas (including 'hot markets' like Austin) for a one family house are $271,000.

The average home price in Austin for a single family residence is $334,000.


https://www.fanniemae.com/singlefamily/loan-limits

Note: Those are conventional financing limits. Anything above that ($417,000 nationwide, with exceptions as high as $625,500 in certain high-priced markets) is considered a "jumbo" loan, and is typically not government/GSE backed.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: